Erdős Problem #1122
Editorial depth: Authored explanatory DRAFT — NOT editorially accepted.
Snapshot classification: open (not automatically the live result).
Subject areas: number theory.
Read this status correctly
The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.
UNACCEPTED DRAFT — mathematical editorial review pending
This material is a provisional, source-qualified manuscript published for line-by-line scrutiny and real HTML/LaTeX testing. It has NOT received separate GCL editorial acceptance or independent mathematical certification. Its presence on this website is not evidence of a solution, reviewed coverage, or final publication.
Review the underlying draft on GitHub.
The question in plain language
An additive arithmetic function assigns a real number to each positive integer and obeys
Suppose such a function almost never decreases when moving from \(n\) to \(n+1\). Erdős Problem #1122 asks whether it must have the simple form \(f(n)=c\log n\).
Here “almost never” means that the set of descents
has natural density zero, or \(|A\cap[1,X]|=o(X)\). The function need only be additive on coprime inputs, not on arbitrary products.
A small example
The function \(\omega(n)\) counting distinct prime factors is additive on coprime products: primes dividing coprime \(a,b\) occur in disjoint sets. Yet it is not a constant multiple of \(\log n\).
For instance, \(\omega(6)=2\) and \(\omega(7)=1\), so \(n=6\) is a descent. A finite exact search finds 65 descents for \(1\le n\le200\) and checks additivity on all 1,547 coprime ordered pairs among the 2,500 candidates \(a,b\le50\).
This finite list does not establish the asymptotic density of descents for \(\omega(n)\); it is not a counterexample to Erdős's conjecture.
Conversely, for \(c\ge0\), \(c\log n\) has no descents. Negative \(c\) makes every step a descent and hence cannot meet the density-zero premise.
What is known
Erdős proved that \(f(n)=c\log n\) when no descents occur, and also under the different assumption that \(f(n+1)-f(n)\to0\).
The maintained catalogue records Mangerel (2022) as proving a partial result when the descent set obeys a stronger logarithmic saving,
and prime values \(f(p)\) satisfy an additional technical growth restriction. Dropping that restriction would misstate the theorem.
The original weaker density-zero question remains open.
GCL's current state
Authenticated audit #2393 records the exact coprime-additive and density quantifiers and the independent finite example. The 1946 and 2022 primary proofs have not been read in full.
What remains
Verify the original Erdős statement and Mangerel's complete prime-value assumptions. Determine whether finite/additive test families illuminate the possible distribution of descents, without converting bounded computations into density theorems.
Sources and claim boundaries
Erdős #1122, last reported update April 2026; Erdős (1946); Mangerel (2022); GCL original-issue return. Published partial results are external. The finite arithmetic is independently checked. Open; unaccepted manuscript.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #2393 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: STATEMENT_CAPTURE_REQUIRED.
- Formal-statement coverage: No statement was confirmed in GCL's pinned formal-provider snapshot. Presence does not mean the statement or proof has been checked.
- External mathematical status: consult current problem page and its dated bibliography.